The 3 Types of AI: A Primer

AI is a booming industry with a lot of noise, thought leadership, and hype. However, as we attend industry events, work with clients, and tell the story of AI for customer data, there’s one common trend. Many of the people we talk to only have a loose idea of what AI does, with little or no mind to what it is.

This post will serve as an easy to read primer on what AI truly is, and what kinds of AI are being developed/where things stand today in the ecosystem.

What type of thing is AI?

Wikipedia has a great breakdown on this which is as follows:

AI is a form of intelligence -> Specifically AI is a synthetic intelligence -> intelligence of a man-made yet real quality.

AI is a type of technology -> A type of emerging computer technology -> A computer technology that performs some intellectual function.

AI is a field of research and study -> It’s an academic discipline which is a branch of computer science.

With this common ground to stand on, let’s look at the types of AI and where they fall.

The 3 Types of AI

Narrow AI

Narrow AI is non-sentient machine intelligence, typically focused on a narrow task.

Narrow AI is the only class of AI that’s currently seeing real world success. For example: AI Chip Saves Google a dozen new data centers, Autonomous swarm of DoD drones, a predictive comedy keyboard trained on scripts, and more. What our company creates (AI driven customer data management & insights) is also an example of Narrow AI. Some products, like your phone, are the product of many narrow AIs working in unison such as voice recognition, predictive text and data management.

Many businesses are at the point where they are sorting out where in their operations it makes sense to apply a narrow AI. Our customers, for example, are at the point where the scale of their data and the complexity of their customer interactions have made traditional hands on approaches to customer marketing unwieldy and expensive. We help them by adding AIs to their data & marketing process to automate and accelerate many of the more intensive data tasks they rely on.

Artificial General Intelligence (AGI) — (hypothetical)

AGI is a machine with the ability to apply intelligence to any problem, rather than a specific and predefined set of problems. It is thought of as being “at least as smart as a typical human”. This does not exist yet, but as we push the boundaries of narrow AI, we’re getting closer.

Defense is one industry looking closely at the feasibility of AGI as a next step from the narrow AI systems already in use. For example, JASON, a DoD advisory group recently published a study on AI feasibility in defense which is a great read if you’re interested in the topic.

JASON says, “AI technologies are of great importance to DoD missions. Defense systems and platforms with varying degrees of autonomy already exist. More importantly, AI is seen as the key enabling technology (along with human-computer interactions of various kinds) of a ‘Third Offset Strategy’ that seeks for the U.S. a unique, asymmetric advantage over near-peer adversaries.”

They go on to say “While many existing DoD weapon systems “have some degree of ‘autonomy’ relying on the technologies of AI, they are in no sense a step–not even a small step–towards ‘autonomy’ in the sense of AGI, that is, the ability to set independent goals or intent,”

One interesting takeaway here is that AI systems replacing actual soldiers in conflict zones would result in an unemployment that’s actually beneficial for those displaced. Other industries, like manufacturers and truck drivers may not be so lucky.

Beyond the technical and engineering limitations we’re faced with, AGI also brings many moral and societal questions that haven’t yet been answered. Would it be right to create and control a new sentient being? Will our goals align? For a great dive into this, listen to Sam Harris’ interview Max Tegmark, an MIT Physicist and AI thinker. This talk reflects on the nature of intelligence, the risks of superhuman AI, the idea of a non biological definition of life, the substrate independence of minds, the relevance and irrelevance of consciousness for the future of AI, near-term breakthroughs in AI, and other topics.

Superintelligence — (hypothetical)

Superintelligence is an AI far surpassing that of the brightest and most gifted human minds.

Where AGI is hardly on the horizon, superintelligence is much more uncertain. A godlike AI seems like a huge leap, but many scientists caution that the moment we unlock AGI, the exponential power of AI could rocket from AGI to Superintelligence. Many warn that this is something that needs to be approached carefully as we barrel down the path of ever more sophisticated AI.

Current AI Trends

Now that you understand a little better what AI is and what types of AI are coming, let’s take a look at exactly where the industry stands today. Gartner published a piece reviewing the current state of AI for business which is very thorough and serves as a great starting point.

The above chart shows various AI applications and where they are in the current business cycle. At ZyloTech, our category is a combination of predictive analytics and several others.

To summarize the research, “A “virtuous cycle” for a new generation of artificial intelligence builds along several critical market dimensions. The risk of ignoring potentially transformational AI exceeds the mitigated risk of fast, early failure.”.

The current market being so weighted to the left shows a period of critical risk/reward where enterprise businesses who wait too long may go the way of Kodak.

The main 3 key conclusions of the research are:

• The risk of failing is great, but the risk of enterprise obsolescence or non competitiveness in the digital business era is even greater.
• While the potential benefits are many, they will come with inherent failures, setbacks, and the “disillusionment” typical of emerging technologies.
• Winners will begin AI programs incrementally — with risk mitigations strategies and techniques in place — but will commit resources for a long game.

For customer marketing at the enterprise level, the time is now to start adopting narrow AI applications strategically. The explosion in data and the complexity in the customer relationship lend themselves uniquely to the horsepower of an AI. The metric driven results of customer marketing are well suited for an AI to be tasked with maximizing.This is why we built an AI that is specific to customer marketer, trained on massive data sources, and incentivized by increasing things like LTV, lowering churn risk, cross-selling and maintaining clean and unified data.

Originally published at www.zylotech.com.